Automated methods for recognizing “named entities” (e.g., persons or places) in a body of text are known. The existing methods have been applied primarily to relatively short works, such as news reports, and highly specialized scientific works such as biomedical texts. Further, these methods have generally been applied to extract and compile information from texts, not to enhance the reading experience.
Written works such as works of fiction often contain a large number of character identities. The character identities and their attributes affect comprehension, interpretation, and understanding of the work and therefore have a profound effect on the reading experience. While most printed copies of written works simply present the work statically in black ink on white paper, the concept of electronic rendering of written works provides an opportunity to customize the presentation of a written work based on the characters within, making the written work more engaging for a user. Existing electronic rendering systems, however, fail to provide such customization.
Corresponding reference characters indicate corresponding parts throughout the drawings.